Agentic BI: Build projects and dashboards from scratch with AI
What does Agentic BI actually look like in practice?
In this demo, we one-shot an entire project from scratch using Lightdash and AI. No prebuilt models, no metrics, no dashboards, and no clicking around in a BI UI.
Starting with raw GitHub data in BigQuery, we use an AI agent (via the Lightdash CLI and MCP server) to:
Generate dbt-style models and a semantic layer
Define metrics for engineering velocity and performance
Create charts and dashboards automatically
Deploy everything end-to-end without touching the UI
This is Agentic BI: AI agents that build, update, and manage dashboards for you, directly from your development workflow.
If you’re a data engineer, analytics engineer, or data leader frustrated with slow, manual BI workflows, this shows what the next generation of BI looks like.
Want to try out Agentic BI?
🔗 Start your free trial → https://www.lightdash.com/start?utm_s...
📣 See how teams use Lightdash → https://www.lightdash.com/customer-st...
📖 Read the docs → https://docs.lightdash.com/introducti...
💬 Join the community → go.lightdash.com/community
#AgenticBI #BusinessIntelligence #AIBI #AnalyticsEngineering #dbt #BigQuery #Lightdash #DataEngineering #AIDashboards